run-behavioral-interview

Run Behavioral Interview

Conduct a structured behavioral interview using STAR probing (Situation, Task, Action, Result) so that candidate assessment is grounded in evidence of past behavior, not self-reported claims or hypothetical answers.

Why This Is Best Practice

Adopted by: Google uses structured behavioral interviewing across all roles, documented in the re:Work hiring guide; Amazon's entire interview process is built around behavioral questions mapped to Leadership Principles using STAR; McKinsey, Bain, and BCG use structured competency-based interviewing; the US federal government mandated structured behavioral interviewing for civil service hiring in the 1980s Impact: Schmidt & Hunter's landmark 1998 meta-analysis (85 years of selection research) found structured behavioral interviews have a validity coefficient of 0.51 for predicting job performance — roughly 2× the predictive power of unstructured interviews (0.38); the improvement comes entirely from standardization: same questions, same scoring rubric, independent evaluation before group discussion Why best: Unstructured interviews — conversational, improvised, rapport-driven — are dominated by the "halo effect" (one strong impression colors all other ratings), affinity bias (interviewers rate similar candidates higher), and first-impression anchoring; hypothetical questions ("what would you do if...") produce polished, socially desirable answers, not evidence; behavioral questions ask for specific past examples because past behavior under real conditions is the strongest predictor of future behavior in similar conditions

Sources: Schmidt & Hunter, Psychological Bulletin (1998); Google re:Work "Structured Interviewing" (rework.withgoogle.com); Campion, Palmer & Campion "A Review of Structure in the Selection Interview" (Personnel Psychology, 1997); SHRM "Behavioral Interviewing" knowledge center

Steps

1. Map questions to competencies before the interview

Each question must map to a specific competency required for the role. Prepare this before any candidate is interviewed — not improvised in the room.

For each competency, write 1–2 behavioral questions:

Competency: Influencing without authority
Question: "Tell me about a time you needed to get buy-in from stakeholders 
who did not report to you. What was the situation, and how did you approach it?"

Competency: Handling ambiguity
Question: "Describe a situation where you had to make an important decision 
without having all the information you needed. What did you do?"

Behavioral question format: starts with "Tell me about a time...", "Describe a situation where...", or "Give me an example of..." — never "What would you do if..."

Cap at 3–5 competencies per interview. Shallow coverage of many competencies produces no signal; deep coverage of 4 competencies produces real evidence.

2. Assign interviewers to competencies (panel interviews)

In panel or multi-round interviews, assign each interviewer to own 1–2 competencies exclusively. This prevents all interviewers from asking the same questions and creates MECE coverage of the role's requirements.

Document the assignment before the first interview in the series:

Interviewer A: Problem-solving, analytical reasoning
Interviewer B: Influencing, communication
Interviewer C: Team collaboration, conflict resolution
Hiring manager: Career trajectory, motivation, values alignment

3. Ask — then probe with STAR

After asking the question, most candidates answer in summary form. Your job is to extract the specific details using STAR probes:

STAR element	What to probe for	Example probe
Situation	Context, stakes, timeline	"What was the situation? What were the constraints?"
Task	Their specific role vs the team's	"What were you specifically responsible for?"
Action	Their personal actions, not the team's	"What did you do? Walk me through your thinking."
Result	Measurable outcome, what they learned	"What was the result? How did you measure success?"

The most important probe: "What did you specifically do?" — candidates often describe what "we" did; STAR requires individual attribution.

If an answer is hypothetical ("I would typically...") redirect: "I'm looking for a specific example. Can you think of a time when you actually did this?"

4. Score independently before discussing

Immediately after the interview, score each competency on a consistent rubric before discussing with other interviewers. A simple 4-point scale:

1 — Strong No Hire: example demonstrates the opposite of the competency, 
    or no specific example provided
2 — No Hire: example demonstrates partial competency with significant gaps
3 — Hire: example demonstrates the competency as expected for this role level
4 — Strong Hire: example demonstrates exceptional competency beyond role level

Write your evidence summary: one paragraph per competency citing the specific example the candidate gave. Do not summarize impressions — cite what they said.

5. Debrief with structured discussion

Bring scores and written evidence into the debrief. Protocol:

1. Each interviewer shares their scores for their owned competencies before any discussion
2. Review the evidence, not the impression: "They gave the example of [specific situation]. Here's how I scored it."
3. Discuss discrepancies: if two interviewers scored the same competency differently, have them share their evidence and reasoning
4. Never anchor on the hiring manager's view first — the manager speaks last, not first, to prevent authority bias

The debrief produces a hiring recommendation, not a vote. If there is meaningful disagreement on evidence, identify whether another interview round is needed.

Rules

• Ask only behavioral questions — no hypotheticals, no brain teasers, no leading questions; past behavior is the only valid evidence type
• Score before discussing — post-discussion scoring is contaminated by social pressure and authority bias
• Probe for individual contribution — "we" answers provide no individual signal; always extract "what did you specifically do"
• Use the same questions across all candidates for a role — changing questions between candidates makes comparison impossible
• Never ask about protected characteristics — age, marital status, national origin, religion, disability, pregnancy; if a candidate volunteers such information, do not probe or record it

Common Mistakes

• Asking "what would you do if...": hypothetical answers reflect imagination, not ability; redirect to "tell me about a time you actually..."
• Accepting vague examples: "I'm generally good at conflict" is not an example; require a specific situation with specific actions and a specific outcome.
• Discussing candidates before all scores are submitted: one strong opinion shared before others submit scores anchors the entire panel; submit scores independently first.
• Asking the same questions across all interviewers: every interviewer probes communication? Then no one covers analytical reasoning; map competencies to interviewers before the process starts.
• Letting the conversation drift to rapport: an enjoyable conversation with a candidate produces no hiring signal; the interview must produce evidence against competencies, not a positive impression.

When NOT to Use

• For very junior or entry-level roles where the candidate has limited work experience — use situational or case-based interviews instead, where the candidate applies their thinking to a presented scenario.
• For highly technical roles requiring demonstrated skill — behavioral interviews complement, but do not replace, technical assessments (code samples, design exercises, case studies); run both.